Search papers, labs, and topics across Lattice.
Affiliation:
3
0
4
Some dialogue games leverage knowledge transfer more effectively than finetuning, with visuospatial tasks showing the highest transfer potential.
A multi-agent framework that leverages diverse VLMs can dramatically enhance zero-shot action recognition by overcoming inherent model biases.
By explicitly guiding attention with predicted action sequences, AGA overcomes the limitations of standard dot-product attention in video action anticipation, leading to better generalization and interpretability.
